© 2004 American Thoracic Society
Hyperglycemia, Bronchial Artery Sclerosis, and Lung FunctionTo the Editor:We read with interest the study by Walter and colleagues (1) about the association between glycemic state and lung function. The authors report a lower FVC and FEV1 in patients with diabetes, whereas the FEV1/FVC ratio was independent of this diagnosis. The diagnosis of diabetes was also associated with a larger reduction in residual (= difference from predicted) FVC than residual FEV1 (1). Consequently, a larger residual FEV1/FVC ratio in patients with diabetes suggested a restrictive ventilatory disorder. Conversely, the association between increased blood sugar and lower than predicted FEV1/FVC ratio was interpreted as an obstructive ventilatory disorder. The authors concede that the overall nature of pulmonary impairment in diabetes remains unclear. The pathophysiologic connection between diabetes and lung function is explained by a possible proinflammatory stimulus of hyperglycemia causing impaired lung function through increased intrapulmonary inflammation and apoptosis. Another possible cause of impaired lung function could be increased sclerosis of bronchial arteries as a consequence of generalized arteriosclerosis in diabetes. Bronchial artery occlusion has been associated with emphysema, fibrosis, and increased intrapulmonary inflammation in experimental studies (2–4). The idea of an intensified emphysema alteration of the lungs after diabetes-induced bronchial artery sclerosis is compatible with the observed pattern of lung function impairment in the study by Walter and colleagues (1). Emphysema-related overinflation can cause severe decrease of FVC accompanied by variable degrees of bronchial obstruction and thus FEV1/FVC ratios. This could explain why no consistent reduction in FEV1/FVC was observed, whereas FVC was clearly reduced in all analyses. Diaphragm elevation with increased closing volume and decreased FVC in absence of detectable bronchial obstruction is another possible cause of impaired lung function. Patients with diabetes had 10 kg more body weight than the control subjects without diabetes in the study by Walter and colleagues (1); however, deficits in the residual pulmonary function and association between glycemic state and lung function were not adjusted for body weight. In light of the common occurrence of patients with diabetes being overweight, elevation of the diaphragm because of intraabdominal fat could be a possible cofactor of decreased FVC and would be compatible with the observed pattern of lung function impairment. Measurement of total lung capacity and residual volume by means of body plethysmography could help to clarify the nature of lung function impairment in patients with diabetes.
Department of Pulmonary Medicine University of Vienna Vienna, Austria REFERENCES
From the Authors: We appreciate the comments of Funk and colleagues and agree that our study does not argue for a particular biologic mechanism linking impaired glucose metabolism to impaired pulmonary function. Arterial disease is a common sequela of diabetes and may contribute to the observed association. Spirometric measures admittedly can be nonspecific; they do not appear to be well correlated with the parenchymal damage of emphysema in a generally healthy population (1). Consequently, a variety of mechanisms linking impaired glucose metabolism to pulmonary dysfunction would be consistent with the Framingham Offspring data (2); the small changes observed in FEV1/FVC ratio do little to differentiate among parenchymal destruction, airways inflammation, capillary bed thickening, or some combination thereof. More extensive pulmonary function testing or histopathologic examination would be necessary. To that end, the Framingham Heart Study has recently begun to measure diffusion capacity, which will enhance our ability to investigate the nature of the pulmonary impairment associated with impaired glucose tolerance. Obesity, which is associated with diabetes, clearly can result in spirometric abnormalities. In an effort to control for this potential confounder, we adjusted for anthropomorphic measures, including weight, in the prediction models. In addition, to ensure that the relation between impaired pulmonary function and glycemic state was not merely the effect of obesity, we explored including either body mass index (weight/height2) or waist/hip ratio, a measure of central obesity; because these had no meaningful effect on the model, only the more parsimonious model was presented. Because adjustment for measures of obesity did not affect the relation of glycemic state to impaired pulmonary function, it is unlikely that the observed associations reflect confounding by body habitus.
Boston University School of Medicine Boston, Massachusetts University of Pittsburgh School of Medicine Pittsburgh, Pennsylvania REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
